Tag: disasters

Another Minsky Moment

Stability leads to instability. The more stable things become and the longer things are stable, the more unstable they will be when the crisis hits. ~ Hyman Minsky

As Dan Alesch has pointed out, we designate disasters by their triggering events, but we remember them for their impacts.  Thus, we know Camille and Katrina and Sandy and Maria and Ian by the devastation they created; had they exhausted their energies over the Atlantic, their names would be forgotten.  In years to come, COVID-19 will cause all of us to shudder, even though we’ve experienced many decades of influenza outbreaks. And it will be hard to forget the 100+ lives lost this week in Texas – “Guadalupe” and “Kerr County” will trigger those memories.

But why were these disasters so impactful?  And, for all of them, why were there such great disparities in those impacts, even among neighboring counties and communities?  My answer – Minsky Moments.

A Minsky Moment is a crisis paradoxically born of stability (It takes its name from Hyman Minsky and his financial instability hypothesis, quoted above).  Minsky believed that a long period of stable financial markets led to ever increasing risk tolerance (and often risk-taking) which in turn led to a sudden collapse in the market.  His ideas have been used to explain both the crisis in Asian markets in the late 1990’s and the Great Recession that we have so slowly climbed out of.

A sad pattern seems to be all-too-frequently repeated. We take action immediately after a disaster and then as its devastation slowly fades from our memories we become more and more tolerant of risk and eventually engage in increasingly risky behavior. Almost invariably, this leads to a Minsky Moment.

For example, the hurricanes in Florida in 2004 and 2005 were the first major storms to hit south Florida since Hurricane Andrew in 1992.  Individually, each was weaker than Andrew, but collectively their impacts were much greater (For example, Wilma – a Cat 3 hurricane – did almost as much damage as the Cat 5 Andrew even after Ivan and Charley had already done so much.).  Over time, people forgot Nature’s devastation – many let their insurance policies lapse; many didn’t properly protect their homes; virtually no effort was made to strengthen buildings built prior to the more stringent building codes put in force after Andrew.  People became so risk tolerant that even common sense precautions (such as properly functioning storm shutters) were ignored.

Craig Colton has pointed out that this behavior happened in New Orleans after Katrina as well.  Homeowners bought insurance in almost record numbers in 2005 and 2006; by 2009, many of those policies had been allowed to lapse.

A very different type of event – a school shooting.  We were all horrified in 2015 by yet another scene of senseless violence, this time at Umpqua Community College in Roseburg, OR.  But the fact that there had been a deadly shooting at Roseburg High School in 2006 was lost in the tragedy of the event.  Clearly the quick response of the police indicates that they, at least, remained highly aware of the risk, but it appears that the leadership of UCC was ill-prepared.  And being unprepared for this kind of an act on a campus in today’s world means that the risk is tolerated and, unfortunately, accepted.

Similarly, even after the devastating effects of the coronavirus in China became apparent (in spite of the Chinese government’s efforts to hide them) in January, 2020, the President and many governors and mayors tried to downplay its potential impacts.  People were encouraged to “party hearty” – Chinese New Year, Mardi Gras and others.  Spring Breakers beached it even into March, and for some, sadly, it was their last Big Wave.  We had not had a major epidemic in 100 years; the false alarms of the last two decades (SARS, Ebola, et al.) conditioned us to believe that the party would never end – until it did.

And just this week, the tragic deaths of campers, and many others, in Texas. The “We can’t afford a warning system” thinking; the “The water in that creek is only a foot deep or so – it won’t flood” mindset; the “We haven’t had a big flood since 1987” excuses led to so many lives wasted.

Perhaps the most touching were Blair and Brooke Harber, two sisters who along with their grandparents, were swept away by the flood. The love of those two sisters – so strong that even the powerful surge of water could not prise their hands apart – is matched only by the weak judgement of the adults who failed them.

Sadly, there can be Minsky Moments in any and every aspect of our lives.  Certainly AIDS took so many lives in the 1980s because of the risk tolerance and risky behavior that were the hallmarks of the sexual revolution and drug use in the ’60s and ’70s.  We had conquered polio in the ’50s; antibiotics seemed able to cure even STDs; there was no real risk – or so we thought.  And yet a virus that apparently had been lying in wait since the ’20s pounced on our risky behavior to become a pandemic. Polio and measles – evil genies we thought we’d eradicated – are again emerging as real threats.

The levees of the Sacramento River Valley provide the basis for a potentially devastating Minsky Moment.  Originally built to provide water for reliable irrigation of farm land, the levee system has led to unrestrained development.  This residential and commercial development is occurring in an area that has seen at least six massive floods (When Leland Stanford became Governor of California, he had to use a rowboat to get to his inauguration.). When the levees breach (one estimate indicates a 64% chance in the next third of a century), the drinking water for 25 million people will be contaminated, millions will be left homeless and tens of thousands will die.  All because we have forgotten the lessons of the Great Mississippi flood of 1927 (John Barry’s Rising Tide provides an excellent explanation of how bad management and engineering contributed to this event. His The Great Influenza is also an excellent reference on the Spanish flu pandemic.).  Similarly, had we remembered the lessons that Nature tried to teach us with the Long Island Express of 1938 (a massive meat cleaver – compared to Sandy’s butter knife – that carved up the Long Island Sound) much of the devastation of Superstorm Sandy would have been avoided.  While some communities had wide beaches and recently constructed berms and dunes that protected them from the worst of the storm, many more of their neighbors went unprotected into that good night.  And those still rage over the blight that is strangling their communities.

In too many cases, the impacts caused by extreme events – especially the human suffering – can be attributed to Minsky Moments like these.  It is all too human to want to forget the bad things that have happened to us.  It is all too human to believe that since no crisis has happened recently, none lies lurking in our future.  But we must go beyond our human failings – we must ensure that fading memories do not give rise to tolerance of risk, then risky behavior, and then the inevitable Minsky Moment. Or else more young voices will go silent, more precious lives will be lost.

Featured

Visualizing resilience

Standing knee-deep in a river and dying of thirst.

Joe Cocker

As I’m writing this, we’re at the end of our election cycle in the US. For months, we’ve been bombarded with snarky snippets aimed at getting us to vote against the other guy, not for somebody. No matter our political affiliation, I think we all sometimes feel we’re in a river of factoids, looking for the truth.

The same thought applies to community resilience. Since I began working in the field, we’ve seen an explosive growth in the knowledge base. Unfortunately, this has not been matched by the application of the knowledge in practice. There are several reasons for this:

  • Accessibility. Much of the knowledge base is captured in academic journals that are never seldom read by anyone other than academics; and even if read, academic jargon and the creep of politics into much of the social science literature turns off many practitioners;
  • Lack of a framework. There isn’t a generally accepted theory of resilience that ties the many disparate strands together;
  • The resilience to __ problem. Practitioners are most often interested in strengthening specific domains and mitigating specific threats, not something as nebulous as fostering a community’s resilience (i.e., practitioners are most interested in the resilience of X to Y). Much of the literature treats resilience as an inherent attribute of a community, ignoring specific threats;
  • Lack of community-specific information. While there are several excellent presentations of data at the state or county level (e.g., Susan Cutter’s maps), there is much less at the level of individual communities;
  • Need to “kiss a lot of frogs.” There is so much information out there (and more being published daily, it seems) that finding that one key paper that will unlock the door to desired solutions requires time and effort that no few practitioners have.

Three years ago, Brian Dabson introduced me to an approach he was developing for the Missouri Transect Project. At the time, I was immersed in the ANCR Benchmarking effort, and – although I praised the overall conception and sent him some suggestions for making it better – I essentially forgot about it. At almost the same time, he left Mizzou for North Carolina (as good an excuse as any to not follow up on my “helpful” suggestions!) and his erstwhile co-workers appear to have dropped the approach as well.

Three months ago, I was asked to consider how to provide meaningful measures for the resilience of small communities, especially in rural areas. I expanded my writ a bit by looking at Opportunity Zones as well. In going back through all of the material I’ve accumulated, I stumbled across Brian’s excellent work. Below, I present my adaptation of Brian’s approach (with apologies to him where I’ve strayed from his original conception). The approach is intended for use by practitioners to determine where to invest scarce community resources.

The concept is deceptively simple. It starts with the concept that the purpose of a community is to carry out common functions for the members of the community. In general, the business of the community – carrying out its common functions – is performed through the consumption and production of community capital – financial, human, social, institutional. Thus, one way to look at a disruptive event is as a disruption of a community’s normal pattern of transactions (thanks due to Dan Alesch for this idea). Recovery then means establishing a new pattern of transactions, i.e., a New Normal. This enables us to assess a community’s resilience in terms of capital – its capital at risk vs the dispatchable capital available for recovery, from a given disruptive event. Examples of fixed and dispatchable assets:

Disruptive events might be natural disasters, or economic crises, or the return of the coronavirus. As discussed in a previous post, the “weaknesses” at the potential point of attack corresponding to the threat comprise the susceptibility. Generally speaking, these are the weaknesses of fixed assets to the threat’s attack. An attractive feature of this approach is that it can be applied to a community system (e.g., housing, water), a neighborhood, or an entire community.

One of the thing that I found very attractive in Brian’s original concept was the way he treated indicators for both susceptibility and recovery. For the Transect Project, he converted each indicator to a value between 0 and 1, by dividing by the range of values. As is generally done, he took the average of sets of indicators to come up with overall values for susceptibility and recoverability. An unintended consequence of this is that this enables us to use qualitative data as well.

For example, if we’re interested in the recoverability of a community’s electric power system, we might have quantitative data relating to financial reserves of its power authority. We might not have quantitative data on its susceptibility to a natural disaster, but through survey data or other means we could come up with a “good, bad, indifferent” rating which we could fuzzify onto a 0 to 1 scale. We then plot recoverability (Y) vs susceptibility (X).

This approach can be usefully applied in several ways. For example, it can be used to look at several threats to determine where to put mitigation dollars. In this figure, I’ve notionally looked at flooding, a health crisis and an economic crisis for a community. For susceptibility to flooding, I would include the condition of houses and other structures, and FEMA flood zone information (for both, there are useful quantitative and qualitative indicators). For recoverability, I would look the fraction of residents living in poverty, whether there were sufficient construction professionals. I would do similar things for the other disruptive events. The results might then look like


In this case, it appears that it might be more useful to invest in mitigating a health care crisis. While there is slightly greater susceptibility to an economic crisis, recovery from a health crisis is much less certain. While recovery from flooding is also “iffy,” a damaging flood is much less likely. Miami provides a real-world example of the latter. Many of the poorer sections of the city (i.e., those with less resources for recovery) are built on higher ground (i.e., less susceptible to flooding).

This approach can be used in other ways as well. For example, flood mitigation funding for Miami might better be used in those low-lying areas with the lowest incomes; i.e., the approach can be used to determine where best to use targeted mitigation money. Similarly, the approach can also be used to determine how to invest. In this case, the different indicators for recovery are compared, as are those for susceptibility. Those that most greatly increase the distance from “red” to “green” are those most likely to have an impact. But since there are costs associated with any action, communities will most likely want to do a “distance / dollar” type calculation. In my next post, I’m going to look at a method a community can use to determine what resources are needed for recovery.

I like this approach for several reasons:
• First and foremost, it is visual. There’s not a lot of numbers or complicated words for the layman to try to understand. If you’re in the red, you want to get in the green.
• Unlike the other common visuals – maps, I can look at how well my community (or my neighborhoods, or my water system…) will handle all of the threats I’m worried about. This makes it easier for a community to prioritize its investments.
• Because I’m looking at all of the community capitals, I can also consider the impact of non-financial investments, and of investments made by all parts of the community. It allows the local government to look at the impacts of investments made by non-profits, businesses, and of “capital stacks” on recoverability. For example, if there were insufficient construction professionals, a partnership could be formed between construction companies, local unions and a community college to begin to fill the need.
• Finally, its extensible. As we learn more about how communities actually recover, and the relative importance of various factors to susceptibility and recoverability, we can add factors or throw out others or learn how best to combine them.

My goal – as always – is to find ways to help communities strengthen themselves. Knowing which strengths are relevant to a community’s ability to withstand or recover from the threats it faces is a crucial first step. That knowledge is the key to taking action to become a stronger – more resilient – community.